SEI Releases AI Engineering Practices for Decision Makers
September 12, 2019 • Article
September 12, 2019—The SEI has released an initial set of recommendations to help organizations that are beginning to build, acquire, and integrate artificial intelligence (AI) capabilities into business and mission systems. “AI Engineering: 11 Foundational Practices,” a white paper written for decision makers, is based on the SEI’s experience assisting government organizations, observing the practices of teams developing and implementing AI and machine learning (ML) capabilities, and learning lessons from industry.
The U.S. Department of Defense (DoD) is making a major push to develop and adopt AI and ML capabilities. The 11 foundational practices just released by the SEI provide initial guidelines to help DoD executives, program managers, and technical leadership ask critical questions in the early stages of building and adopting AI technologies and capabilities. The practices cover choosing the right problems for AI to solve, selecting the appropriate solutions, managing change in AI and ML technology, and others.
Drawing on its history of establishing a software engineering discipline and its present work in creating AI solutions for mission, the SEI is planning the development of an AI engineering discipline to further help the defense and national security communities develop, deploy, operate, and evolve mission capabilities that leverage rapidly evolving AI and ML technologies. Future SEI publications will detail the components of AI engineering and the SEI’s plan to establish it as a discipline.
“These foundational practices offer some initial insights in a chaotic landscape where AI and ML technologies are evolving and advancing rapidly,” said Matt Gaston, director of the SEI Emerging Technology Center. “An AI engineering discipline can help the DoD create and adopt solutions that are reliable, reproducible, trustworthy, and maintainable.”
In the near term, the SEI is forming a community of interest around AI engineering for defense and national security. Collaborators, customers, stakeholders, and subject matter experts will work together to understand and set the course for AI engineering. Those interested in being part of the community of interest should email the SEI at firstname.lastname@example.org.
To download “AI Engineering: 11 Foundational Practices,” visit https://resources.sei.cmu.edu/library/asset-view.cfm?assetid=633647.